Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[MDM,]

compasspnt wrote on Fri, 10 October 2008 09:08

OK. If one did a mix through a desk (Protools > analogue desk / outboard > capture)... And then did the same song, mixing fully ITB... And then you did a null test (of course they would not null)... What exactly would the leftover difference tell you...?

that you think too much?

[Ross Hogarth]

ok my 2 cents
and
i have tried to remain neutral

I pretty much hate the sound of the digi 192 i/o
it is smeared and the low end is bunched up
it is a necessary evil often for large tracking sessions but for all my overdubs i use a cranesong 192 hedd
now
as far as taking and converting my whole mix out of the digi 192 d/a
it is completely unacceptable to me
i refuse to reconvert out of that box and THEN a/d back in
no thank you
i do have my digi 192 hooked up to outboard gear so when i want to compress through a piece of hardware i can or use my dmx or 2016 or whatever outbaord
but
my mix goes d/d aes to my hedd 192 for some tape and pentode color but goes back in aes to PT and gets recorded on an audio track
i monitor the dig out of pt d to d into my avocet so my system completely ignores the analog out of protools
i find that the analog out of protools sounds nothing really like my actual mix
i am sure you guys like and hear a difference with your dangerous 2 buss but you will not get me to use that until it becomes an aes box ignoring the a/d and d/a of the digi192


here is the simple routing

[Tomas Danko]

Ross Hogarth wrote on Sun, 12 October 2008 02:36

ok my 2 cents and i have tried to remain neutral I pretty much hate the sound of the digi 192 i/o it is smeared and the low end is bunched up it is a necessary evil often for large tracking sessions but for all my overdubs i use a cranesong 192 hedd now as far as taking and converting my whole mix out of the digi 192 d/a it is completely unacceptable to me i refuse to reconvert out of that box and THEN a/d back in no thank you i do have my digi 192 hooked up to outboard gear so when i want to compress through a piece of hardware i can or use my dmx or 2016 or whatever outbaord but my mix goes d/d aes to my hedd 192 for some tape and pentode color but goes back in aes to PT and gets recorded on an audio track i monitor the dig out of pt d to d into my avocet so my system completely ignores the analog out of protools i find that the analog out of protools sounds nothing really like my actual mix i am sure you guys like and hear a difference with your dangerous 2 buss but you will not get me to use that until it becomes an aes box ignoring the a/d and d/a of the digi192 ]

Hi Ross,

Say, won't the external D-A and A-D conversion through the HEDD 192 be excluded in ProTools' automatic delay compensation, forcing you to nudge the incoming audio tracks manually?

Regarding a Dangerous 2 Buss unit with AES/EBU I/O only, that would lock the user down to that one choice of digital audio conversion. Instead of, say, being able to choose Apogee 16x's, Prism ADA8 or a bunch of Lavry Blue's. A great sounding 2 Buss digital I/O unit would probably be very cost prohibitive, and so a sellable product could end up being somewhat of a compromise regarding the conversion quality (which, as far as I know, stands totally against anything Dangerous does).

Cheers,

Danko

[Ross Hogarth]

respectfully, i could care less about saleability and the dangerous 2 buss
i just don't want my stems converted with the digi 192
and if they were coming out aes
then the converters used by dangerous would have to be ..the shit
and
to make myself clear
the way i use my system is for ITB mixes
my mixes are recorded at the session sample rate and are not going to a analog tape machine
when i was mixing on a console i never recorded back to PT except as a safety
I always went to tape and maybe dsd or some other super high res dig format
as far as latency
the hedd has basically no latency
no artist i have ever worked with has ever noticed
and in the mix setup it is acting as a mix processor
similar to what a mastering engineer would use it for
and
when i do want to use my alan smart c1 compressor on the 2 buss
then and only then does my mix get converted out d/a and a/d
and that is through the hedd routing which is super flexible for that

[Tomas Danko]

I see, you're sending the final stereo mix through the HEDD 192 and your analog outboard gear, then back into the computer the same way. Regardless of how small or large the latency is through the HEDD 192, it really doesn't matter in this case. It's all serial processing, so it'll work totally fine.

[Ross Hogarth]

yes
and very often these days
i am not analog processing the mix
only hedd processing which is d/d
the only analog processing would be an alan smart c1 or my 33609 or maybe a 2254 with some eq
but i am finding that no d/a - a/d is clearer

[Extreme Mixing]

Thanks for sharing your working methods Ross.  I too have stayed out of this for the most part, except to say that those of us who have work to do inside the box will find ways to make it sound good, while others will scratch their beards and comment on how much better it could be, if only...

Steve

[wwittman]

compasspnt wrote on Fri, 10 October 2008 10:08

OK. If one did a mix through a desk (Protools > analogue desk / outboard > capture)... And then did the same song, mixing fully ITB... And then you did a null test (of course they would not null)... What exactly would the leftover difference tell you...?

That you need to get out more...

[Blas]

This place and this thread is sooooooo addictive.  Last week my mother-in-law passed-away and we had to go up to Chicago over the last 5 days.
Never had time (and didn't think it too proper) to jump on my wife's family computer to check things.  So as soon as we hit home again Sunday evening, before unpacking the car, guess where I headed to?  Right here.  Just to keep up with this ongoing discussion.
Wife yelled in "where did you go? The car's still loaded!"  Just checking my email...I can be such a liar.

Blas

[compasspnt]

Joe,

Glad you're back.

Sorry to hear about your Mother in law.

[Blas]

Thanks Terry.

The lady was one month away from hitting 90.  She had a lot of good years, which apparently runs in my wife's family.  I lost my father at age 71.  Makes me want to make sure I enjoy all the life God gives me.

Anyway, it just is always so interesting how EVERYONE here at the forums live and breathe the desire to create better and better audio.  We're always looking for that perfect cup of coffee.  And the search will never end, which makes for good conversation between friends.

Joe

[Peter Weihe]

wwittman wrote on Wed, 15 October 2008 18:09

compasspnt wrote on Fri, 10 October 2008 10:08 OK. If one did a mix through a desk (Protools > analogue desk / outboard > capture)... And then did the same song, mixing fully ITB... And then you did a null test (of course they would not null)... What exactly would the leftover difference tell you...? That you need to get out more...

Yes, that's exactly what I felt when I had finished the 0-test today.
I had performed a few summing tests for a German studio magazine and a huge test of line level stages for the University of Hannover over the last years so it was a small step to do the o test Bill had asked for.
As I had expected, the 0-tests did not show anything but that each audio stage and each converter the signal has to pass through changes the sound of the signal in it's own way. Once you send the signal through more than one stage it's impossible to tell what stage is responsible for what part of the leftover. And as every analogue mix of digitally recorded signals must pass through DA converters and a summing stage there are too many variables to let you recognize the difference the analogue summing process makes in comparison to the digital summing.
Anyway I will post some of my findings in my next post.
The results are probably too theoretical to help anybody to make better mixes but they may show how ridiculous the level of most of the typical digital vs analogue summing debates is.

[Bill Mueller]

Peter Weihe wrote on Fri, 17 October 2008 19:17

wwittman wrote on Wed, 15 October 2008 18:09 compasspnt wrote on Fri, 10 October 2008 10:08 OK. If one did a mix through a desk (Protools > analogue desk / outboard > capture)... And then did the same song, mixing fully ITB... And then you did a null test (of course they would not null)... What exactly would the leftover difference tell you...? That you need to get out more... Yes, that's exactly what I felt when I had finished the 0-test today. I had performed a few summing tests for a German studio magazine and a huge test of line level stages for the University of Hannover over the last years so it was a small step to do the o test Bill had asked for. As I had expected, the 0-tests did not show anything but that each audio stage and each converter the signal has to pass through changes the sound of the signal in it's own way. Once you send the signal through more than one stage it's impossible to tell what stage is responsible for what part of the leftover. And as every analogue mix of digitally recorded signals must pass through DA converters and a summing stage there are too many variables to let you recognize the difference the analogue summing process makes in comparison to the digital summing. Anyway I will post some of my findings in my next post. The results are probably too theoretical to help anybody to make better mixes but they may show how ridiculous the level of most of the typical digital vs analogue summing debates is.

Peter,

Excellent! Thanks for doing this, no matter what the outcome.

Best regards,

Bill

[Peter Weihe]

Bill Mueller wrote on Fri, 10 October 2008 16:41

what I meant was to take an ITB mix, route it both to the two buss and split it to a level matched analog summing console. Then record the outputs from both back to the DAW and compare them. I'm not talking about using an SSL with outboard reverbs and trying to match an ITB mix. From what I gather, this has been done already and I missed it or did not have time to look into it. If anyone has any more info about these kinds of tests, I would appreciate it and might actually have time now to listen to the results. I don't believe I have a dog in this fight, because of the rig I use, but I often wonder if a dedicated digital console like mine and PT or MX rig constitute ITB? Personally, I doubt it, because I am spreading the processing load across double the processing capacity with the digital console. Thanks, Bill

Hi Bill,

some years ago when the dangerous 2 Bus first appeared on the market my technician Manni and I were asked to perform the requested summing test for a German studio magazine.
Test 1
We used a Pro Tools session recorded in Abbey Road 1 performed by The Royal Philharmonic Orchestra. We simply used the monitor mix of the recording engineer and tried to copy the volumes and panning on three different analogue summing boxes or consoles.
First we bounced a 1 Khz sine wave and copied it to each track in the Pro Tools session. Then we soloed each track with the sine wave and measured the levels on each side of the master with the waves PAZ meter. The next step was to send the tones out of 16 individual outputs of Apogee 16x DA converters to the summing boxes and send the stereo output back through Apogee AD converters to a Pro Tools stereo track. We adjusted the master level control of the summing box until the PAZ meter on the input channel showed the same levels as in the original session. We had to fine tune some of the Apogee's output levels with the Dangerous 2 Bus. That was of coarse much easier with the console 's faders.
One was my 1976 Helios console and the other a simple passive mixer that Manni had built for this session with stereo faders used as symmetrical faders directly connected to a discrete symmetrical summing stage via summing resistors and left and right switches. The purest mixing board I have heard so far.
Each mix was clearly  sounding different from each of the others while all of the three analogue mixes had one aspect in common. The dynamic was significantly different from the digital mix. That was not only easy to hear but also easily measurable. I told the founders of Logic about our test and they came to my studio with their top development engineers and after they had heard the mixes they asked me for the files.

Beside the aesthetic differences which are obviously a matter of taste there were those dynamical differences:
The Pro Tools mix showed the lowest levels and the highest peaks. It was the most dynamic mix regarding the difference between the quietest and the loudest signal.

But the analogue mixes sounded louder. Obviously all of the analogue mixes looked and sounded slightly compressed in the quiet parts. Consequently reverb tails of the majestic Abbey Road 1 hall seemed to be more detailed in the analogue mixes.
In addition the waveforms looked as if there was a peak limiter on the loudest fast percussive signals.
That is especially funny because marketing people of analogue summing devices obviously love to praise the extra headroom of their boxes. It seemed to us that they mix it up with compression.
The most dynamic analogue mix most similar to the Pro Tools mix came out of Manni's Mixer.

However while the pp parts sounded as if they were brought up by a compressor in the analogue mixes, the ff parts sounded as if somebody pushed the master fader up and the Pro Tools Mix sounded thinner in these parts. The ff parts were clearly louder in the analogue mixes and that's what the waveforms showed.

We had the impression that the Pro Tools mix engine was not able to sum all tracks
to their full dynamic level while the peaks were still higher. Listening to the main mics (decca tree with 3 M50 ) in solo however showed this dynamic.

The analogue mixes sounded more satisfying to all of us, rockers and classical Tonmeisters. A master compressor, a peak limiter a slight broad lift in the lower midrange and riding the levels at the ff-parts brought the Pro tools miy closer to the analogue mixes.

I am not saying which mix was closer to truth.

[Peter Weihe]

Test 2

When the heads of the rock and pop department of the University Of Music, Hannover  asked me to help them building their new studio I played the results of the summing test. It was obvious that the design of the summing stages made a huge difference sound-wise. In order to find the the mixer they were looking for they asked me to isolate the variables from each other and reduce a new test on the pure sound-differences of different line stages.

I chose some solo instruments and some mono monitor mixes of sessions I had played on and each of them helped to reveal certain aspects. Then I send them out through Lavry Blue DA converters and recorded them back through Lavry Gold AD.
That was my DA/AD check.
Now each candidate was inserted in the Lavrys if possible carefully leveled and recorded. With some units like API eqs which alway rise the level by 1-1,5 dB we had to lower the output of the DA converter.
We collected everything the market has to offer plus all my vintage units and included them in our test. API 82oo-7800, V72, V74a, Tonelux, Chandler TG, BA 1084, Great River , different 1:1 transformers, SSL ( so called superanal..) blah blah.....

Over the years I have recorded every unit that happened to come to my place and meanwhile there are 96 tracks. It's a test and a half.

When I read Bill's post I thought, that I could quickly make the O-test with some of these files but then....

I realized that it was not possible to get the 44,1 KHz files perfectly in phase.
Even moving the files one sample I could not get them perfectly in phase because they were recorded slipped by a quarter or 1/2 sample.
That was the case with all files recorded with the Lavry Gold converters.

I opened a new session and repeated the DA/AD check with PT 192 converters.
Surprisingly they made it possible to get the recorded  files in phase with the original file and with the other newly recorded files each time.
But- the 0-Test ( one file out of phase) of the the original and the DA/AD Check file
sounded like a joke!
The leftover was -25dB. That's how close the extra DA/AD path 192 i/o sounded to the original.
However performing the 0-test with 2 AD/DA Check files ( that means 2 equally re-recorded tracks) almost perfectly nulled. There was no audible leftover.
- 75.5 dB
That means that Pro Tools is able to record two times absolutely sample accurate
and that the huge difference between the original and the path through the 192 i/o must be blamed to their sound.
I should add that almost all tracks of the the original mono mix had been recorded through Lavry Gold converters and once you have heard those tracks listening back throughPT 192 you don't want to ever leave the Box again.
I absolutely agree with Ross.
Maybe the difference would have been much less dramatic, if the files had already been recorded through the PT 192 converters as then the sound quality would have been diminished ( in comparison to first class converters) before.
On the other hand our Abbey Road Orchestra file was recorded with Pro Tools converters and the difference to the extra DA/AD path was still obvious.

My conclusion is that the DA/AD path alone makes such a big difference to the original sound that it would be impossible to analyse what extra difference would be
due to analogue summing or analogue stages when performing a 0-test digital vs analogue summing.

All those statements come to my mind where people on forums believe in the wonder of analogue summing without taking the quality of the converts into consideration.

Try it again with the new Cranesong Egret DA converters with 8 channel mixer and then there will be a real difference.

BTW. The discrete class A stage in the  Cranesong Avocet ( the same as in the Egret) was the most neutral line stage in our test. That's an incredible unit.

The guys from the University Hannover liked the line path through the Great River EQ the best followed by the API rack mixer.